读取文件时如何格式化每一行？

Question

Input format:

2014-09-21 00:09:22,718 TRACE [user: admin12] common.Log (PerformanceExtractor.Python:9776) - ClientId:895,UserId:258,Ip:111.1.1.1,DurationMls:23,DurationString:0.023 seconds,Url:Calculate.LoanExmple

My output would be into variables; for example:

$date = 2014-09-21 00:09:22,718 $user = admin12 $ClientId= 895 $UserID=258 $ip = 111.1.1.1 $time=0.023 $url=Calculate.LoanExmple

In JAVA I would tackle this using a loop, a Stream Object, and a regular expression. I have no idea how to tackle this using Perl. I will also insert this variable as a column into a database and the line will be at least 3000 max 5000 each time i will launch the .pl.

my loop is

{
print $line;

--formatting here?

last if $. == 500;

}

This just prints out the line as above - I guess the best solution would be to format it and get the values into variables while reading each line, ready to INSERT into DB with DBI libraries.

any suggestions?

Answer 1

Something like this?

while (<$fh>) {
    my @fields = m{^
        (\d{4}-\d{2}-\d{2}\s\d{2}:\d{2}:[\d,]+)
        \s TRACE \s
        \[user:\s(\w+)]
        \s common.Log \s \(PerformanceExtractor\.Python\:\d+\) \s - \s
        ClientId:(\d+),
        UserId:(\d+),
        Ip:([\d.]+),
        DurationMls:\d+,
        DurationString:([\d.]+) \s seconds,
        Url:(\S+)
    $}x
        or next;  # skip lines which don't match regexp

    printf('$date=%s; $user=%s; $client_id=%s; $user_id=%s; $ip=%s; $time=%s; $url=%s', @fields);
    print "\n";
}